Home

Version 2.0 Beta

Develop and analyze high-performing seasonal streamflow forecasts using PyForecast, developed by Reclamation's Great Plains Region. PyForecast takes advantage of multi-threading and multiple processor cores to analyze thousands of forecasts in minutes using cutting-edge statistical techniques.

Table of Contents

• Users Manual
  • Installation & Updates
  • Software Overview
    • Summary Tab
    • Stations Tab
    • Data Tab
    • Forecast Options Tab
    • Regression Tab
    • Density Analysis Tab
  • PyForecast Best Practices
  • Example Forecast Development
• Scientific Background
  • Datasets
    • Snow Water Equivalent (SWE)
    • Air temperature & Precipitation
    • Streamflow
    • Soil Moisture
    • Climatic Indices and Teleconnections
  • Model Creation
    • Brute Force Selection
    • Parallelized Sequential Selection
    • Multiple Linear Regression
    • Principal Components Regression
    • Z Score Regression
    • General Regression Neural Network (In Development)
    • Gaussian Process Regression (In Development)
    • LASSO Regression (In Development)
    • Data Transformations
    • Data filling and interpolation
  • Uncertainty and Forecast Distribution
    • Normal Distribution
    • Student's t Distribution
    • Log-normal Distribution
    • Generalized linear model (GLM)
    • Composite distribution using kernel density estimation
  • Hindcasting
  • Forecast disaggregation
    • Analog year
    • Quantile mapping
    • Downscaling using k-nearest neighbors
• Appendices
  • PyForecast Data Loaders